Media preference increases polarization in an agent-based election model

Abstract

Western societies have become more polarized over the last decades which forms a threat to their democracies. It is therefore important to understand the detailed mechanisms behind this polarization in the framework of opinion dynamics. Recent work has emphasized the role of the people’s interactions with (mass) media in driving polarization, in particular through the formation of echo chambers. Here, we study how these echo chambers emerge from the collective behavior of people within a social network in the presence of media. For this, we use a new agent-based model of the election dynamics in a two-party system. In this model, media are highly connected and influential nodes, which are randomly located in the network and have the role of spreading external influence (e.g. information on the state of the economy) throughout the population. The model, with properly tuned parameters can reproduce the overall properties of US election results, together with the representation of numerous details, such as the portion of non-voters. Echo chambers emerge in this model through a media-preference feedback, when voters preferentially surround themselves with media that have their political opinion. In this way, the model provides valuable information on how polarization arises through collective behavior of people and media.